acetylene chemistry - American Chemical Society

---

.

ACETYLENE CHEMISTRY W. E. HANFORDl AND D. L. FULLER General Aniline and Film Corporation, Easton, Pa. Acetylene chemistry has taken rapid strides, especially in Europe, during the past twenty years as a result of the extensive research and development by Reppe and his co-workers of I. G. Farbenindustrie. These advances were due primarily to the fact that they found how to handle acetylene at 200 to 300 pounds per square inch at temperatures up to 200' C. This was accomplished in general by diluting the acetylene with an inert material, such as nitrogen, or by constructing suitable apparatus. Data on the explosive behavior of acetylene and certain mixtures are presented. Two types of reactions of acetylene have been developed: reactions where the product resulting from a reactant and acetylene seems to be best explained by migration of an acetylenic hydrogen, and those which can best be explained by addition of a reactant to the acetylene linkage. Topics discussed are vinylation, ethynylation, carboxylation, and cyclic polymerization.

A

CETYLENE chemistry was one of the major contributions of the German scientists to their war effort. United States scientists also were actively engaged i n trying to make acetylene chemistry a real contribution of organic chemistry to the American cause and did add several small parts from this field to the Allies' war effort. Because this is an extremely broad subject this paper is limited to the f i s t member of the acetylene series, CaH2. The chemistry of the oxidation of acetylene is excluded, as are the conventional methods for the manufacture of such compounds a8 vinyl acetate, divinyl acetylene, ethylidene diacetate, tetrachloroethane, trichloroethylene, and tetrachloroethylene. Each represents a major development in the field of chemistry that has been practiced in this country for many years. The reactions of acetylene discussed below are not as well known, but many believe they will become as important in the future development of organic chemistry in this country. They differ from those excluded in that in the majority of cases the acetylene must be used under pressure in order td obtain a reaction rate fast enough for commercial operation. A great deal of the information given was taken from the reports obtained from the I. G. Farbenindustrie by technical missions sent to Germany and also from data obtained in the authors' laboratory (1,3,6,16, I!?). The development of this field is due to a very large extent to the work of Reppe and his associates a t the main I. G. Farbenindustrie research laboratory a t Ludwigshafen, Germany. This paper may be looked upon, in general, as experiences a t General Aniline and Film Corporation with these reactions. HANDLING OF COMPRESSED ACETYLENE

Because the development of new techniques for handling acetylene under pressure of several atmospheres is the factor that has made these new reactions possible, before the chemistry is considerkd, a discussion of these pressure techniques is in order (8, 4).

decomposition of acetylene, but unfortunately this reaction can and does in many cases go so rapidly that the result is a detonation. Under these conditions, the I. G. has reported instantaneous pressures up to 20 times that calculated for the straight thermal decomposition or over 200 times the initial pressure. The characteristics of acetylene have not changed HAZARDS. as a result of any research. What has changed is a better understanding of the factors that enter into the control of the tremendous force that can be developed during its decomposition. I n the General Aniline laboratory an explosion test was made

on a nitrogen-acetylene mixture in an extra heavy pipe carrying a right-angle bend, Because of interest in the safety and design of acetylene feed lines, the test was carried out a t room temperature with an 80 to 90% acetylene mixture and nitrogen a t 175 pounds per square inch gage in a 1-inch extra heavy flanged steel p i p 6 feet long. This pipe had a working pressure of 1361 pounds m t h a safety factor of 8. At one end, the explosion was set off by fusing a n i r o n wi.re; the flange a t the other end held a steel plate 0.130 inch thick which had been used many times in static strain gage tests with nitrogen a t 2000 pounds per square inch without taking a permanent set. In view of the current rule that tenfold pressure increases occur when acetylene mixtures explode, this plate was expected to be satisfactory for the test. When the explosion occurred the 10,000-pound gage connected by st steel tube '/le inch in insfde diameter did not have time to move appreciably. The report was loud, more like a shot un than a rifle, and there was an instantaneous bright yellow fight (of burning carbon as it hit the air) in the barricade. There was no damage, no windows or light bulbs were broken, and the pipe held, but a 1.25-inch hole was cleanly punched out of the steel plate. This slug knocked chips out of the concrete floor, SO I t came out at high speed. If we assume the ultimate shearing stress of the plate t o be 42,000 pounds per square inch, we find the shearing force to be 21,500 pounds and the pressure on the slug t o be 17,500 pounds per square inch. This is about ten times the tenfold pressure increase expected. The effective pressure then was considerably greater than that given by the usual rule. The uantity of acetylene present in this pipe was about 10 grams &igure I).

'

Thermodynamically, acetylene is unstable with respect to its elements a t room temperature and when decomposed to its elements a t 18' C. liberates 53,500 calories per gram mole. As the usual decomposition products are mainly carbon and hydrogen, the rise in pressure must be due to an increase in temperature. On this basis, and assuming no heat loss, the temperature would increase from 15" to around 3100' C. and thus result in a pressure of some twelve times the initial pressure. This theoretical calculation checks very well with the experiment$alresults that have been obtained for the straight thermal 1 Present address, M. W. Kellogg Company, 225 Broadway, New York, N. Y.

This force surely belongs t o the top group of explosives. However, it is possible to run acetylene reactions under pressure without hazard, i f the proper precautions are taken a t all times. Two general procedures have been worked out for controlling the.hazardous behavior of acetylene: (1) diluting the acetylene with some inert gas, such as nitrogen or carbon dioxide; and (2) keeping the amount of free space present in the system t o a minimum by using bunches of small tubes and by filling all large voids with a packing material, such as steel or porcelain Raschig rings (Figure 2). Both methods have their particular uses, depending on the reaction which is to be carried out. For autoclave operation, the first method is probably the brst;

1171

INDUSTRIAL AND ENGINEERING CHEMISTRY

1172

I\

BARRICADE

KWALL CzHz -Nz

SUPPLY

Figure 1.

Acetylene Explosion Test in Bent Pipe

Since the authors were initially interested in carrying out these reactions of acetylenr in autoclaves, they employed for the explosion tests a 1200-cc. Aminco bomb, 3 inches 111 diameter, which had been pressure-tested at 6000 pounds. (The volume of the corresponding reactor used by I. G. for this study m-ah 730 cc.) As with most gas explosion reactions, the size and shape of the reaction vessel have ~1 large effect on the pressure developed. The presence of a diluent in the acetylene, however, greatly reduces the possibility of detonation regardless of the size and shape of the bomb. As the temperature of acetylene is increased, the pressure a t which i t will decompose is drcreased (Table 111). The results of the decomposition of mixtures of acetylene with hydrogen, acetylene with nitrogen. and acetvlene with ethvlene are summarized on Figures 3 and 4. The areas below the curves represent safe operating conditions and those above the lincs represent unsafe operating conditions. I n the running of organic reactions in autoclaves in batch operations, another constituent is always present in the gas phase-the vapor of the liquid or diluent used in the reactor. This means that with volatile solvents the reaction conditions in the autoclave are safer, as far as detonation is concerned, than those that exist in the acetylene supply lines. However, the acetylene will be safe in the lines, provided the diameter is less than 0.5 inch. Before all the I. G. data became available the authors studied many of these reactions and their approach to the problem was a very practical one. If they had decided to run, say, the vinylatiori of methanol with a mixture of 35y0 acetylene and 65% nitrogen at 250 pounds gage pressure and 180" C. with an excesb of methanol present at all times, they would ask: Is this mixture safe or can it decompose under these Conditions? The bomh mentioned above was loadrd with the desired reaction mixture -

for continuous operation, especially where a solid contact catalyst is to be used, the second method is preferred. I n determining the eGplosive characteristics of acetylene, some means must be selected for initiating the decomposition. A great deal of thought and experimental work has been expended in deciding on the best method for initiating this reaction. Consideration has been given to a spark plug, fusion of a wire, and dynamite caps. The fusion of a wire has been used by all investigators in determining the explosive characteristics of acetylene. In attempting to select ignition material, the I. G. studied the critical pressure for decomposition of pure acetylene with variouq kinds of wires (Table I). TABLEI. LIMITING PRESSURE FOR DECOMPOSITION OF ACETILENE AS 4 FUNCTION OF MELTING POIKTOF WIRE Melting Point,

c.

Element

2550

N O

Pt Fe cu A1 Pb

1760 1530 1080 660

330

Pressuie Lb./Sq In , 'Ab? 20.6 20.6

25.2 27.0 38.9

110.0

Vol. 40, No. 7

I

TABLE 111. LIMITINGPRWSEREFOR 'DECONPOSITION OF ACETYLENEAS A FWWTION O F TEMPERATURE O F GAS A T T I h r E ?F IGNITION OF T T 7 1 ~ ~ Temperature,

From these data i t is apparent that the lover the temprraturr of fusion of the wire, the higher the pressure can be before decomposition occurs. Other factors which must be considered are the diameter of tlir wire, the length of the wire, and the rate a t which the energy is put into the wire., The I. G. decided to use platinum wire 0.5 mm. in diameter and 10 em. long fo rall decomposition studies. In their work the authors have used iron wire, rrhich gives slightly higher values for the pressure than are obt,ained with platinum. Explosive decomposit,ions of this type also depend upon the dimensions of the vessel. The I. G. studied the effect of the size of the vessel on the decomposition of pure acetylene under pressure (Table 11) and concluded that the critical pressure for. decomposition is independent of the diameter for vessels greater than 7.9 inches in diameter, and that it is very markedly a function of diameter for smaller vessels.

c.

16

100 150 180

Pressure, Lb./Sq. In., Abc;. 20.6 18.4 16.6

15.6

TABLE11.

L I N I T I X G PRESSURE FOR DECOMPOSITIOK O F ACETYLENE AS A FUXCTION O F DL4METER O F VESSEL

Vessel Diameter, Inches 2.0 3.9 5.9

7.9

10.6

Pressure, Lb./Sq. In., Abs. 26 5 23.5 21.3 20.6 20.6

CROSS SECTION

Figure 2.

Pipe Packing to Prevent Acetylene Detonation

I N D U S T R I A L A N D E N G I N E E R I N G CHEMISTRY

July 1948

1173

The I. G. concluded from the results that pure acetylene can be handled safely at pressures up to 15 atmospheres absolute, provided the large pipes are filled with bundles of tubes having diameters of 0.5 inch or less and that the free spaces at bends, etc., are filled with Raschig rings. Under these conditions thermal decompositions did occur, but no detonations; provided detonations can be avoided, safe plant designs are feasible. As the Reppq synthesis of butynediol from acetylene and formaldehyde operated at pressures of 5 to 10 atmospheres, usually 5 to 6, ample margin of safety was afforded by building the unit for 100 atmospheres operation. For long continuous runs the bundles of tubes are preferred to the Raschig ring packing. As for the chemistry of acetylene under pressure, there are four important. types of reactions of acetylene: vinylation ethynylation, carboxylation, and cyclic polymerization. VINY LATION

For the vinylation of alcohols (12, 16-18,26, 28) with acetylene, a bas? is used as the catalyst. For this reaction, potassium HC=CH

PRESSURE OF MIXTURES,

PSI.

and an attempt made to decompose i t by fusing an iron wire in the gas phase. If no reaction occurred, the operation was repeated once or twice. The acetylene pressure was raised and the operation repeated two or three times more. Finally the decision was made on the basis of data of this type whether or not the experiments were justified. Mixtures used in these testjs and the conditions are summarized briefly in Table IV.

TABLE IV.

CONDITIONS OF TEST

Other Component Pressure Present in Liquid Temp., Lb./Sq. I;., % CzHa % N t and Gas Phases C. Abs. 70 30 151 Cyclohexane 235 165 67 33 285 Cyclohexane 50 98 50 205 Cyclohexane 98 38 62 205 Cyclohexane 50 50 , 151 260 Cyclohexane 0 120 100 205 Methanol 100 0 120 215 Methanol 71 29 120 Methanol 215 33 67 120 135 Butanol 100 0 120 Isobutanol 85 100 0 120 245 Vinyl isopropyl ether Neglecting vapor pressure of organic liquids, Compositions

Explosion Yes No Yes NO

NO No Y e8 No No Yes No

+ RON +H,C=CHOR

hydroxide or potassium alcoholate gives better results than any of the other alkalies tested. The general reaction conditions required for the vinylation of alcohols from methyl to octadecyl are: addition of 1to 5% of potassium hydroxide to the anhydrous alcohol; heating to a temperature of 160' to 185' C.; then treating with a mixture of 1 part of acetylene and 2 parts of nitrogen at a total pressure of 100 to 300 pounds gage. The method for carrying out this reaction in a batch operation is as follows: A stainless steel autoclave provided with an efficient stirrer and capable of withstanding a pressure of 2000 pounds is charged with the alcohol-say, anhydrous butanol-and to this is added 23% of anhydrous potassium hydroxide. The autoclave is closed, evacuated, and flushed with nitrogen which contains less than The flushing o eration is continued until the ni0,2y0oxyg!n. trogen leaving the autoclave sgows less than 0.5% oxygen when tested. The autoclave is heated to 160"C. and the nitrogen pressure adjusted to about 130 pounds gage. Acetylene is then introduced through a regular high pressure feeding system until the pressure in the autoclave is 200 pounds gage. Under these conditions the acetylene is absorbed rapidly and the drop in pressure is made up by the addition of fresh acetylene. I n the authors' laboratory acetylene at 250 pounds pressure is generally used directly from cylinders. The reaction proceeds

It is surprising how accurately this type of experiment can be run. Using data of this type, work in the field of acetylene under pressure has been carried on €or over four years without any serious accidents; there was only one rupture of any equipment during these four years and this was in one of the feed lines. The seaond method of preventing the detonation of acetylene by keeping every acetylene molecule within a certain maximum distance from a solid such as a wall has been studied in great detail by I. G. Following the preliminary work on the diameter of vessels, etc., the I. G. decided to investigate the explosion characteristics of acetylene on a commercial plant scale. These experiments were conducted by burying pipes of different sizes, from'0.5 to 8 inches in diameter and over 100 feet long, in deep trenches and covering them with earth and heavy ties. Instruments were installed for determining the pressures at several points along the pipes. I n these pipes were placed various types of steel plugs or arrestors, bundles of small tubes running the complete length of the pipe, bundles of small tubes with voids at the ends, and similar bundles of tubes in which the voids a t the ends were filled with Raschig rings, etc. These assemblies were filled with acetylene mixtures and decomposition was initiated by fusion of a platinum wire.

PRESSURE OF

MIXTURE, Ps.1.

1174

INDUSTRIAL A N D E N G I N E E R I N G CHEMISTRY

Figure 5.

Flow Sheet for Vinylation of Alcohols

smoothly. The autoclave is checked from time to time to see that the reaction mixture does not contain more than 33% acetylene at any time during the run. When reaction has ceased, the autoclave is cooled, the acetylene and nitrogen are vented, and the product is worked u p by distillation. Under the conditions outlined, one can obtain yields of 80 to 95% based on the alcohol with all of the primary alcohols from methyl to octadecyl. With the secondary alcohols, such as isopropyl or see-butyl, vinylation occurs but the yields are lo-wer. The reaction does not proceed nearly as well p i t h tertiary alcohols, such as t e r t butanol. Polyhydric alcohols, such as ethylene glycol, glycerol, sorbitol, and sugar derivatives, can also be vinylated under these same conditions. The products of polyhydroxy compounds are generally divinyl, trivinyl, etc., derivatives, or the cyclic acetals or mixtures of the two.

Vol. 40, No. 1

drogen. This, of course, is what one would anticipate, for when potassium butylate is heated to a temperature of about 185 C., potassium butyrate is formed. The control of this side reaction is an important variable in the successful commercial operation of this process for two reasons: (1) -4s the catalyst is consumed, the rate of reaction is reduced; this, of course, adds to the COST of the operation. (2) The hydrogen concentration in the inert gas and unconverted acetylene recycled gradually builds up and some method of removing i t or venting i t from the system muqt be used. The vinylation of alcohol has also been studied in the vapoi phase, using potassium hydroxide on calcium oxide as a contact catalyst. The life of this catalyst is short and the extreme difficulty in removing the heat of reaction makes the liquid system more attractive. The vinyl ethers so prepared are colorless liquids or solid. which can be utilized for the preparation of a wide variety 01 organic chemicals and polymers. They can be polymerized ai low temperature with acid catalysts, such as boron trifluondt., and interpolymerized with a variety of unsaturated compound. by means of a peroxygenated catalyst such as benzoyl peroxidc The double bond in the vinyl ethers is extremely reactive and undergoes all the characteristic reactions of a carbon-carbo11 double bond. Vinyl nitrogen compounds, such as vinyl carbazole (17, 2.5 aiid vinyl diphenylamine, can be prepared by the direct reactioii of the amine with acetylene in the presence of potassium hydroxide. HC=CH

+

q-'?-(j-i",

"/

G \ NH/ C i

I

HC=CHz

Here the reaction conditions are essentially the same, except that in those cases where the amine is high melting, a solvent is generally employed. The easiest to prepare and most stablp of thp vinyl compounds of this series is vinyl carbazole.

This was DreDared in cvclohexane with potassium hidrdxide and zi"nc oxide as cataC H 2 0 C H ~ C H 2 lyst a t 180" C., 200 pounds gage pressure, HC-CH 1 +I and using a mixture of 2 parts of nitrogen I CHzOH CH~OH 4- L H 2 - d C H C H s + CH20CH=CH2 with 1 part of acetylene. The vinylation is rapid, but the duration of the reaction deOther functional groups can also be present in the alcohols, pends to a large extent on the degree of provided they do not react with potassium alcoholate at 160' C. agitation. When acetylene is no longer absorbed, the reaction mixture is allowed to cool. The unreacted carbazole and the It is more difficult to vinylate the lower alcohols, such as inorganic materials are removed by filtration, the cyclohexane methyl and ethyl, than the higher alcohols by a batch operation is distilled, and the vinyl carbazole purified by distillation, or because of the high vapor pressure which results at the elevated by crystallization from such materials as cyclohexane, benzene, temperatures. The lower members can best be obtained by and methanol. The yields are above 85%. carrying out the operation in a continuous fashion, although the Diphenylamine can be vinylated under the same conditil Iris, same equipment can be used for the higher vinyl ethers. The but the yield is lower and tlie product is extremely sensitive to air way in which the reaction is carried out in a continuous manner and difficult to obtain in a pure form. Vinyl carbazole is readill is shown diagrammatically in Figure 5 , polymerized by acid- or peroxide-type catalysts to produce polyvinyl carbazole. It can also be polymerized by thermal The tower, A , is charged with the anhydrous alcohol and the potassium hydroxide or alcoholate as catalyst. On the top of A methods. It has been used for eIectrical insulation because of is a column, B, to separate the vinyl ether from the unused alcoits high electrical resistance and low power factor. It is sold hol, acetylene, and nitrogen. The nitrogen diluent is brought under the tradc-mark of Polectron and its main war use was in down to about 15 pounds pressure, acetylene is added, and the making condensers for the proximity fuse. mixture of acetylene and nitrogen is passed through the compressor to obtain the desired pressure and then into the bottom of A . Because the less basic nitrogen compounds in the amine clash Continuous operation is maintained by adding make-up alcohol vinylate readily, i t is natural to suppose that certain amides will by means of pump C. The synthesis of methyl vinyl ether by also vinylate easily. The compound that has been studird i 1 1 this process is carried out at about 300 pounds gage pressure, due greatest detail is the cyclic amide, pyrrolidone (11). to the vapor pressure of methanol at 160" to 185' C. %-Butanol, on the other hand, can be vinylated at 100 pounds pressure and HzC-CH2 octadecyl alcohol from atmospheric pressur; to about 100 pounds.

+

CHzOH

CHzOCH=CHz

CHr-O\

The only side reaction that occurs to any appreciable extent in this reaction is the gradual conversion of potassium alcoholate to the potassium salt of the corresponding fatty acid and hy-

For the vinylation, it is best to form first the potassium salt of the pyrrolidone. This can be done by mixing potassium hy-

droxide with pyrrolidone and distilling out about one third of the pyrrolidone, which carries with it the water from foremation of the potassium pyrrolidone: This reaction mixture 1s then charged into an autoclave, heated to about 160 " C. and treated with dilute acetylene at 200 pounds per square inch gage. The absorption is very rapid a t first, then gradually dies off. TJsually vinylation reaction is not run to completion, because the rate drops off very rapidly near the end. A 70% conversion seems t o be a good point t o stop. At this time the reaction mixture is discharged and the vinyl pyrrolidone boiling a t 95" c. a t 14mm. pressure is separated from the unchanged pyrrolidone. This material can then be polymerized by either a peroxide catalyst 01: a boron trifluoride catalyst to a solid material, which is readily soluble in water with the formation of a very viscous solution.

.

Polyvinyl pyrrolidone has been used in Europe as a synthetic blood plasma and also for offset printing. How far these developments will go in this country is still to be determined. The higher acids, such as lauric, stearic, and oleic, can be vinylated with acetylene by using the potassium or zinc salt as catalyst under essentially the same conditions as those employed for the vinyl ethers (90).

8

0

ClrHaa(!

- OH + HC=CH +CIJH~I, - OCH=CH2

The best yields are obtained when the system is anhydrous. Vinyl stearate is readily obtained from a reaction mixture of 5% zinc stearate and 95% stearic acid heated in an autoclave at 165O C., using diluted acetylene a t 200 pounds pressure. Vinyl stearate resulting from this reaction mixture is purified by distillation. It can be polymerized under about the same conditions as are used for the lower molecular weight vinyl esters. Potassium stearate is also a good catalyst for this reaction. Although it is extremely difficult t o vinylate acetic acid with potassium acetate as a catalyst, the potassium salt is a good catalyst for the higher fatty acids. Relative solubilities of the salts may be part of the answer. The same general reaction conditions can be employed for the preparation of vinyl laurate, vinyl oleate, etc. It has been reported that the polyvinyl esters of the unsaturated acids of high molecular weight have interesting drying characteristics. Phenols, especially the monoalkylated phenols, react with acetylene very readily in the presence of organic bases and selected zinc salts, such as zinc naphthenate, a t approximately 160" C. and 200 pounds total pressure t o give resinouq materials. This is the basis for the synthesis of the rubber tackifier developed in Germany known as Koresin and now offered by General Aniline and Film Corporation (91). The specific product is prepared by mixing p-tertbutyl phenol with about HC=CH -+10% by weight of zinc naphthenate, heating the reaction Hac-C-CHa mixture to 160" C., ClHa

iYY

1175

INDUStRIAL AND ENGINEERING CHEMISTRY

July 1948

+

CHr OH I

/

and introducing a mixture of 2 parts of nitrogen and 1 part of acetylene t o a total pressure of 200 pounds. The reactants are kept under these conditions until a resinous material with certain predetermined physical constants has been obtained. The degree of polymerization in this reaction is not large, being of the order of 7, and the product contains many products including some unchanged alkyl phenol. Practically all the monoalkyl phenols give a similar product under the same conditions. There has been a great deal of speculation by Reppe and others as to the mechanism of f o r m s tion and the structure of this resin. It seems probable that the hydroxyl of the alkyl phenol reacts with acetylene to form a substituted vinyl phenyl ether. Under the conditions of the reaction however, this vinyl ether reacts like any other vinyl ether with a phenol to give a resin which probably has an ethylidine structure. Reppe, in his formulation of Koresin, has given it the alternative ethylene structure. Because infrared absorption shows the presence of CHI and OH groups and nearly the same type of material can be made from p-tertbutyl phenol and acetaldehyde or p-tert-butyl phenol and a vinyl ether, this formulation does not seem possible, It would be interesting to prepare the vinyl tert-butyl phenyl ether and heat it under the conditions used in the vinylation to see if polymerization of this type will occur. Mercaptans (thiols), according t o the Reppe patents, form vinyl thioethers. The reaction proceeds under approximately the conditions outlined for the other active hydrogen compounds. I n general, all the active hydrogen compounds which will form stable potassium salts-alcohols, amines, acids, phenols, and mercaptans-will react with acetylene under pressure a t 160" C. to give in the majority of cases good yield of vinyl derivatives. What is the mechanism of these reactions? So far, no basic data have been published, so guesses will need verification. With phenols and alcohols the following series of reactions looks reasonable. KOH

+ ROH

ROK

+ HOH

+= OR- + K + H + ORHC=COR-

ROK HC=CH H HCFZCOR-

+ ROH

CHz=CHOR

+ OR-

A similar set of reactions appears suitable for acetylene and amines such as carbazole, or with phenols and acetylene with amines as catalysts. I n the case of acetylene and aqueous trimethylamine with no added catalyst, we can write a number of untested mechanisms, such as

+ HOH HCCHz+ + OH(CHJsN + HC=CHz+ + OH- +[(CHa),NCH=CHd+ HC=CH

OH-

ETHYNY LATION

In addition to its unsaturated characteristics, acetylene possesses active hydrogen atoms which should add to unsaturated systems. This reaction is called ethynylation. Sodium acety-

1176

INDUSTRIAL AND ENGINEERING CHEhISTRY

lide and the Grignard derivative of acetylene will react with a carbonyl group to form an acetylenic alcohol. Nieuwland found that both the unsaturated nature and the hydrogen of acetylenr could be activated by an aqueous mixture of potassium chloride, ammonium chloride, cuprous chloride, and hydrochloride acid to form monovinyl acetylene and higher polymers. Reppe has found another copper salt, copper acetylide, which will catalyze the hydrogen addition reaction to several types of unsaturated systems, especially aldehydes and ketones. Because the only details that have reached this country from I. G. on this general reaction are for the synthesis of butynediol-1,4 and propargyl alcohol, the present discussion is limited to the reaction condi. tions used for the preparation of these two products (2, 6, 7 , 9. 13-1 8, 22-24). HCECH HC=CCHsOH

+ CHZO +HCZEECCH~OH + CHZO +HOCHZC-CCHeOH

The catalyst for these reactions is prepared by impregnating silica withcopper nitrateand bismuthnitrateand heating to 500 "C. to convert thc nitrates into a mixed oxide; the best one contains at this step about 12% copper and 3% bismuth. The mixture is packed into a tower and a formaldehyde solution and then acetylene a t 5 to 6 atmospheres is passed down the tower to generate the catalyst, which is a copper acetylide-formaldehydeacetylene-water addition complex. When on stream a 15% formaldehyde solution and pure acetylene are passed over this catalyst a t 100" to 120" C.; it is important to keep the temperature below 130" C. This control can be accomplished by evaporating some of the water or by using reactors of small diameter. As catalysts go, a t least to a petroleum chemist, the copper acetylide is relatively inactive and requires a long contact time, from 0.33 t,o 3.5 hours, depending on the recycling system and the percentage of unchanged formaldehyde that can be tolerated in the crude product. The reaction mixture is filtered to remove any copper acetylide which may have come into the reaction product, and the reaction product is then distilled. The men who have seen the operation in Germany and those who have operated i t in this country have found that the operation is not nearly so bad as i t sounds, if proper equipment is used. As pure acetylene was used in this reaction, the acetylene was handled under pressure by the second method outlined above --that is, bundles of tube. By the hydrogenation of butynediol, butenediol and butanediol are produced. The latter formed the basis of a butadiene synthesis via tetrahydrofurane, and it also can be converted to butyrolactone via dehydrogenation and then to pyrrolidone with ammonia. All these materials and many others were used to produce a great variety of compounds for practically all the fields of organic chemistry. Carbonyl compounds reported to react in a similar mannei with acetylene are acetaldehyde, propionaldehyde, n-butyraldehyde, dodecanal, benzaldehyde, and acetone. Both the monoand di- addition products have been reported. Another reaction which was found earlier by Reppe but has not been studied in great detail is the reaction of meLhylolamine$ with acetylene in the presence of copper acetylide ( 1 7 ) . Thr reaction of dimethylmethylolamine and acetylene with coppei acetylide as the catalyst is as follows:

Vol. 40, No. 7

Both the mono- and di- addition products have been isolated. Similarly, dimethyl01 urea and trimethylol melamine react with acetylene to form addition compounds, the structures of which are not accurately known because they have never been isolated. On the basis of analogy, one would formulate the reaction as follolrs: F-IOCHzNHCNHCHzOH

II

+ HC-(>H

---+

0 '

+

HOCH~NHCNHCH~CZZZCHHOH

I1

0 CARBOXY LATlON

Another interesting reaction which has been developed b> Reppe and his eo-workers is the synthesis of acrylic acid derivatives from acetylene (8, 16). Apparently Reppe was initially interested in trying to make propargyl aldehyde by the combination of carbon monoxide and acetylene in the way in which he made propargyl alcohol, by the combination of acetylene and formaldehyde. He studied the reaction of nickel carbonyl and acetylene in the presence of acids. Instead of obtaining propargyl aldehyde, he obtained acrylic acid as the main product of the reaction.

A gas, acetylene, reacts in a 2-phase liquid system with both components as rapidly as the acetylene can be added and the temperature kept under control. A 500-cc. 4-necked flask was equipped with a condenser, stirrer, thermometer, and bubbling tube. The flask was charged with 96 grams of methanol and 128 grams of nickel carbonyl. During vigorous agitation, 133 grams of concentrated hydrochloric acid and 90 liters of acetylene (25 C. and 760 mm., washed with water and dried) were added during 3 hours. T i t h an ice bath the temperature was held a t 25" to 30" C. At the end of this time, 40 liters of acetylene were added (during 30 minutes) as long as there was liberation of heat. About 300 cc. of water and 100 cc. of ether were added and the two layers were separated. I n the reaction 181.3 grams of methyl acrylate (78.7%theoretical based on nickel chloride), 4 grams of methyl propionate, and 86.8 grams of nickel chloride were formed The reaction ran as well as an Organic Syntheses preparation. The method as outlined here requires the use of a stoichiometric quantity of nickel carbonyl. Reppe found that it is possible to convert the nickel chloride back to nickel carbonyl' by the reaction of nickel chloride with ammonia and then treatment of the nickel ammonium chloride with carbon monoxide.

Ni(NH3)sCL

+ 5CO + 2H20 --+ Ni(C0)4

+ 2NHaCI + (NH4)&03+ 2Pu'H1

Whether or not this synthesis will compete with some of the other syntheses that have Seen and are being developed in this c?untry, remains to be seen. Regardless of its economics or commercial importance, it is an extremely interesting reaction and a worth-while addition to our knowledge. Reppe has found that if the alcohol contains water, hydrogenation of the double bond of the acrylic acid does not occur; but if the system is anhydrous, dehydrogenation occurs and the yield of the acrylic ester is lower. The same reaction is stated t o go

INDUSTRIAL AND ENGINEERING CHEMISTRY

July 1948

for all the alcohols and also for the synthesis of amides by replacing the alcohol with ammonia and amines. Reppe believes that the formation of cyclopropenone occurs by the direct combination of carbon monoxide and acetylene, and this is then opened through the addition of the active hydrogen of the alcohol t o the system to give acrylic ester. This mechanismis substantiated by the fact that substituted acetylenes give both isomers expected by a ring opening. Another very interesting reaction in this same field is the reaction of carbon monoxide with acetylene in the presence of cobalt carbonyl hydride, H.Co (CO)r, to yield hydroquinone.

1117

its higher homologs which is more likely to be soluble in the reaction medium. If one wishes to run an ethynylation with acetylene, the best catalyst will probably be copper acetylide, some other copper compound, or some other acetylide. For the addition of carbon monoxide to acetylene, the only effective catalyst which has been reported to date is nickel carbonyl. There is a wonderful field here for academic study of the nature of the catalysts used in these acetylene reactions, and also for obtaining further information on the mechanism and cont(ro1 of the decomposition of acetvlene. ACKNOWLEDGMENT

OH

I

2HC=CH

+ 3CO + HzO --+

Q

+ COz

CYCLO-OCTATETRAENE

One of the reactions of acetylene developed by Reppe which has probably caused more discussion than any other is the direct formation of cyclo-octatetraene from acetylene under pressure (10,37). H H

. 4HC=CH

C=C

/

Hd -3

\

CH

HJ

AH

\

/

c=c

‘H H The catalyst for this reaction IS nickel, cyanide, which is prepared by the reaction of nickel chloride with hydrogen cyanide. The catalyst is suspended in a selected solvent, anhydrous tetrahydrofurane, and is charged to the reactor along with a small amount of calcium carbide which is added to maintain anhydrous conditions. The direct polymerization of acetylene then occurs a t a temperature of 60 O to 70 O C. and with an acetylene pressure of about 250 pounds. Under slightly higher pressures and somewhat higher temperatures] some of the higher vinylogs of cyclo-octatetraene are obtained. Their structures, however, have not yet been elucidated. After the acetylene has been absorbed, the catalyst is filtered from the reaction mixt8ure,and the cyclo-octatetraene is isolated by distillation. The physical properties of cyclo-octatetraene obtained by this process agree with those previously repoxted by Willstiitter. Numerous reactions of cyclo-octatetraene have been studied by I. G. chemists, and it has been found to undergo some very unusual transformations. It can be hydrogenated to a hexahydro derivative and then oxidized by means of nitric acid to gfve a good yield of suberic acid, shqwing that the ring hexahydro derivative is an eight-member ring. This compound will undoubtedly be the subject of many papers within the next few years. SUMMARY

Omitted from this discussion are acetylene reactions such as reduction of acetylene to ethylene, addition of hydrogen fluoride, hydrogen sulfide, hydrogen cyanide, and ammonia. For work with acetylene under pressure some means must be selected of determining whether the reaction conditions to be used are safe or unsafe. To run a vinylation with acetylene] the best catalyst to test first is a potassium salt in an anhydrous medium. If this is not satisfactory, then the zinc or cadmium salt of the active hydrogen compound should be tried, or one of

The authors wish to thank their associates in the laboratory for their assistance. LITERATURE CITED

(1) Alt, “Acetylene Chemistry,” U. S. Dept. Commerce OTS, PB L60,285 (Mar. 28, 1947). (2) Appleyard, C. G. S., and Gartshore, J. F. C., “Manufacture of Butynediol a t I. G. Ludwigshafen, Germany,” Ibid., PB 28,556 (Aug. 9, 1946). (3) Callott, W. S., “Researches in Acetylene Chemistry at I. G. Farbenindustrie, Ludwigshafen, Germany,” Ibid., PB 9626 (April 19, 1946). (4) Copeland, N. A,, and Youker, M. A,, “German Techniques for Handling Acetylene in Chemical Operations,” Ibid., PB 20,078 (July 26, 1946). (5) Evans, D. C., “New Technical Applications of Acetylene.” Ibid., PB 18,953 (June 7, 1946). (6) Fuller, D. L., et al., “Manufacture of Butynediol (and Certain Related Materials) from Acetylene and Formaldehyde,” Ibid., PB 80,334. (7) Hopkinson, R., et al., “I. G. Farbenindustrie, Synthetic Rubber Plant at Ludwigshafen, Germany,” Ibid., PB 1763 (Feb. 15, 1946). (8) I. G. Farbenindustrie, Modern Plastics, 23, 162 (1945). (9) I. G. Farbenindustrie, “Preparation of Butynediol,” U. S. Dept. Commerce OTS, PB 19,320 (June 7, 1946). (10) Kammermeyer, K., “Cyclo-octatetraene,” Ibid., PB 62,593 (May 30, 1947). (11) Kline, G. M., Modern Plastics, 23, 157 (1945). (12) Ibid., 24, 159 (1947). (13) Monrad, C. C., “Reppe Butadiene Proccss,” U. S. Dept. Commerce OTS, PB 4604 (Mar. 8, 1946) (14) Niemann, G., “State of the Reppe Process,” Ibid., PB 50,458 (Feb. 28, 1947). (15) Reppe, J. W., “Acetylene as a Basis of a New Industrial Chemistry,” Ibid., PB 2437 (Feb. 8, 1946). (16) Reppe, J. W., “Advances in Acetylene Chemistry,” Ibid., PB 1112 (Jan. 25, 1946). (17) Reppe, J. W., “Advances in Acetylene Chemistry at I. G. Ludwigshafen,” Ibid., PB 13,366 (May 24, 1946). (18) Reppe, J. W., “Butynediol ap a Basis for Plastics Intermediatez,” Ibid., PB 11,394 (April 5, 1946). (19) Reppe, J. W., “Synthesis of Intermediates for Polyamides on a n Acetylene Basis,’’ Ibid., PB 25,553 (Aug. 2,1946). (20) Reppe, J. W., et al., U . S. Patent 2,066,075 (Dec. 29, 1936) (21) Ibid., 2,072,825 (March 2; 1937). (22) Ibid., 2,300,969 (Nov. 3, 1942). (23) Ibid., 2,319,707 (May 18, 1943). (24) Ibid., 2,232,867 (Feb. 25, 1941). (25) Rose, J. D., “Products Formed by the Interaction of Acetylene and Amines,” U. S. Dept. Commerce OTS, PB 25,667 (Aug. 16, 1946). (26) Schildknecht, C. E., et al., IND. ENG.CHEM.,39, 180 (1947) ; 63 references given. (27) Textile Research Institute, “Synthetic Fiber Developments i n Germany,” U. S. Dept. Commerce OTS, PB 7416 (March 29, 1946). (28) Zoss, A. O., and Fuller, D. L., “Oppanol C,” Ibid., PB 67,694 (May 30, 1947). REC~IVED March 26, 1947. Presented before the Division of Organic Chemistry, Symposium on Some Contributions of Organic Chemistry t o the War Effort, a t the 110th Meeting of the AMERICAN CHEMICALSOCIETY, Chicago, Ill. L